公开(公告)号：US20240095276A1

公开(公告)日：2024-03-21

申请号：US18151288

申请日：2023-01-06

Applicant: Shutterstock, Inc.

Inventor： Raúl Gómez ,  Alessandra Sala

IPC: G06F16/535 ,  G06F16/583 ,  G06F16/587

CPC classification number: G06F16/535 ,  G06F16/583 ,  G06F16/587

Abstract: A method for recommending results to a user from a search query is provided. The method includes receiving, in a search engine, a search query for a media file from a user, identifying a style preference of the user associated with a one or more media file attributes, based on a user-related search history, selecting, from a database, a one or more media files based on the search query and the style preference of the user, determining a style preference score for the one or more media files based on the media file attributes, and recommending to the user a top ranked media file based on the style preference score. A system including a memory storing instructions and one or more processors to execute the instructions to cause the system to perform the above method is also provided.

公开(公告)号：US20240112217A1

公开(公告)日：2024-04-04

申请号：US17937342

申请日：2022-09-30

Applicant: Shutterstock, Inc.

Inventor： Emily Teoh ,  Diarmaid Finnerty ,  Lauren Sarah Burnham-King ,  Veronica Darling ,  Alessandra Sala

IPC: G06Q30/02 ,  G06F3/0482 ,  G06V20/30

CPC classification number: G06Q30/0242 ,  G06F3/0482 ,  G06V20/30 ,  G06V2201/10

Abstract: Methods and systems for predicting performance of creative content are disclosed. Exemplary implementations may: receive a collection of images; provide a context to a user; serially cause display of pairs of images on a computer interface; receive user responses indicating which image of each pair is preferred given the context; determine a resonance value for each image based on a number of times the user responses indicate each image is preferred when displayed in a pair of images; determine a confidence score for each image; generate one or more models for predicting image performance based on one or more of the resonance value and the confidence score for each image; receive a plurality of candidate images; determine, using at least one model, a first metric set for each candidate; and cause display of a listing of the candidate images, the listing including the first metric set for each candidate image.

公开(公告)号：US20250148004A1

公开(公告)日：2025-05-08

申请号：US19012597

申请日：2025-01-07

Applicant: Shutterstock. Inc.

Inventor： Raúl Gómez Bruballa ,  Alessandra Sala

IPC: G06F16/535 ,  G06F16/583 ,  G06F16/587

Abstract: A method for recommending results to a user from a search query is provided. The method includes receiving, in a search engine, a search query for a media file from a user, identifying a style preference of the user associated with a one or more media file attributes, based on a user-related search history, selecting, from a database, a one or more media files based on the search query and the style preference of the user, determining a style preference score for the one or more media files based on the media file attributes, and recommending to the user a top ranked media file based on the style preference score. A system including a memory storing instructions and one or more processors to execute the instructions to cause the system to perform the above method is also provided.

公开(公告)号：US20250139155A1

公开(公告)日：2025-05-01

申请号：US19012588

申请日：2025-01-07

Applicant: Shutterstock, Inc.

Inventor： Raúl Gómez Bruballa ,  Alessandra Sala

IPC: G06F16/535 ,  G06F16/583 ,  G06F16/587

Abstract: An embodiment includes determining a style preference score for a new media file based on a recommendation model, wherein the new media file is selected by a user; and when the style preference score is lower than a pre-selected value: identifying a new attribute in the new media file that is not included in the model, updating a first coefficient in the recommendation model based on a value of the new attribute, scoring multiple media files in a user-related search history of the user with the model to form a ranked list based on the updated first coefficient, and updating a second coefficient in the recommendation model when the new media file is not at a top position of the ranked list, relative to the multiple media files in the user-related search history, and storing the updated recommendation model in a training database.

公开(公告)号：US12210565B2

公开(公告)日：2025-01-28

申请号：US18151288

申请日：2023-01-06

Applicant: Shutterstock, Inc.

Inventor： Raúl Gómez ,  Alessandra Sala

IPC: G06F16/53 ,  G06F16/535 ,  G06F16/58 ,  G06F16/583 ,  G06F16/587

Abstract: A method for recommending results to a user from a search query is provided. The method includes receiving, in a search engine, a search query for a media file from a user, identifying a style preference of the user associated with a one or more media file attributes, based on a user-related search history, selecting, from a database, a one or more media files based on the search query and the style preference of the user, determining a style preference score for the one or more media files based on the media file attributes, and recommending to the user a top ranked media file based on the style preference score. A system including a memory storing instructions and one or more processors to execute the instructions to cause the system to perform the above method is also provided.

公开(公告)号：US20240394730A1

公开(公告)日：2024-11-28

申请号：US18324408

申请日：2023-05-26

Applicant: Shutterstock, Inc.

Inventor： Raúl Gómez ,  Alessandra Sala

IPC: G06Q30/0201 ,  G06Q30/018 ,  G06Q30/0601

Abstract: A method for providing contributor recommendations to users of an online content marketplace is provided. The method includes retrieving an attribute of a user of an online content marketplace, identifying contributors of the online content marketplace based on the attribute of the user, scoring user-contributor pairs according to a dense embedding of the attribute of the user and a dense embedding for each of the contributors, providing, to the user, a list of the contributors ranked according to the score of the user-contributor pairs, receiving, from the user, a contributor selection, and providing, to the user, multiple content files from a gallery of the selected contributor, for use in a media application running on a client device with the first user. A system including a memory storing instructions and a processor to execute the instructions to cause the system to perform the above method are also provided.

公开(公告)号：US12106548B1

公开(公告)日：2024-10-01

申请号：US18618982

申请日：2024-03-27

Applicant: Shutterstock, Inc.

Inventor： Raúl Gómez Bruballa ,  Alessandra Sala

IPC: G06V10/774 ,  G06N3/0475 ,  G06V20/70

CPC classification number: G06V10/774 ,  G06V20/70

Abstract: A method for training a generative image model, including defining multiple sensitive categories and protected attributes associated with multiple training images, determining for a particular sensitive category a distribution of a protected attribute within the training images, and based on the distribution, calculating for each training image a corresponding image debiasing weight value associated with the protected attribute. The method further includes generating annotated training data including the training images, and for each training image, (1) the corresponding image debiasing weight value associated with the protected attribute and (2) a corresponding descriptive text caption. The method further includes performing a training process using the annotated training data to train a generative image model resulting in a trained model. A contribution of each image in the training images to an optimization loss of the training process is weighted during the training process using the corresponding image debiasing weight value.